Memorial Sloan Kettering, City College of New York Awarded $4M NIH Grant to Study the Use of Machine Learning in Early Breast Cancer Detection


Memorial Sloan Kettering Cancer Center and the City College of New York (CCNY) were recently awarded a $4 million grant from the National Institutes of Health to study the use of machine learning for early breast cancer detection in high-risk women. The estimated five-year project co-led by Elizabeth J. Sutton, MD, Associate Member in MSK’s Department of Radiology and a breast imaging specialist, will harness the power of machine learning to help the team analyze medical images to determine those at greatest risk, as early as possible, while limiting the burden of screening in this population.

Over the years, clinicians, engineers, and physicists from both institutions have collaborated and identified this area of study where artificial intelligence has the potential to result in clinical implications. Together, the team will analyze 100,000 breast exams from MSK, Duke University, and Johns Hopkins University. Experts from MSK and CCNY believe the large volume of data gives them a unique opportunity to identify future risk, including being able to detect some breast tumors earlier.

“We’re hopeful the results will enable us to take steps towards precision screening in the future where we can determine the required breast MRI screening interval for each individual at increased risk of breast cancer,” said Dr. Sutton.

Risk will be estimated from magnetic resonance images (MRI) of the breast as well as mammograms, using deep learning techniques, a type of machine learning that emulates learning in the human brain through many layers of processing. A retrospective analysis of the large dataset will determine if some women could have avoided unnecessary scans without missing newly developing cancers.

“Women with a strong family history or related genetic mutations have an elevated risk of breast cancer and are recommended to participate in yearly MRI screening,” said Dr. Sutton. “However, the rate of detection in this high-risk cohort is small, prompting a desire to reduce unnecessary MRI exams.”

The project, “Machine learning for risk-adjusted breast MRI screening,” will be co-led by Lucas C. Parra, Harold Shames Chair and Professor of Biomedical Engineering at the CCNY.